US20050096859A1 - Noise measurement system and method - Google Patents
Noise measurement system and method Download PDFInfo
- Publication number
- US20050096859A1 US20050096859A1 US10/699,526 US69952603A US2005096859A1 US 20050096859 A1 US20050096859 A1 US 20050096859A1 US 69952603 A US69952603 A US 69952603A US 2005096859 A1 US2005096859 A1 US 2005096859A1
- Authority
- US
- United States
- Prior art keywords
- intermediate frequency
- noise
- frequency signals
- signal
- measurement
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R29/00—Arrangements for measuring or indicating electric quantities not covered by groups G01R19/00 - G01R27/00
- G01R29/26—Measuring noise figure; Measuring signal-to-noise ratio
Definitions
- One measure of the integrity of a communication system is the amount of noise present within the system. Characterizing the noise of a communication system involves measuring noise power within one or more specified frequency bands, shown for example in the exemplary noise spectrum of FIG. 1 . Measuring noise power, in turn, involves measuring or estimating the statistics of the noise, which is inherently random.
- a conventional noise figure meter (shown in FIG. 2 ) provides accurate measurements of noise power within a designated frequency band.
- noise within the frequency band is down-converted and measured by a noise power detector. Because the noise figure meter includes multiple frequency conversion stages, the noise figure meters are typically expensive to manufacture.
- a direct conversion receiver is typically less expensive to manufacture than a noise figure meter because the receiver includes only a single frequency conversion stage.
- the direct conversion receiver converts the noise within a frequency band (for example, frequency band c shown in FIG. 3C ) to a baseband noise signal (shown in FIG. 3D ) that is measured by a noise power detector (as shown in FIG. 3A ) or by a narrow-band analog-to-digital converter (as shown in FIG. 3B ).
- these direct conversion receivers are not as accurate as noise figure meters.
- One measure of error, the variance of the measured noise power is substantially higher for the direct conversion receiver than for the noise figure meter.
- the variance of the noise power measured by a typical direct conversion receiver is approximately twice as great as the variance of the noise power measured by a noise figure meter.
- FIG. 4A An alternative approach to noise power measurement is shown in FIG. 4A .
- a frequency conversion stage causes upper and lower noise sidebands to overlap within a single measurement band B.
- FIGS. 4B-4D show these overlapping noise sidebands within the single measurement band B as a local oscillator within the frequency conversion stage is stepped in frequency between frequencies f 1 , f 2 and f 3 .
- a noise power detector measures the noise power in the single measurement band B, with the local oscillator at each of the stepped frequencies. While this approach can take advantage of low-cost signal processing to extract the noise power in a designated frequency band (for example, frequency band c) based on the overlapping noise sidebands, measurement accuracy is not as good as that of the conventional noise figure meter. For example, the variance of the noise power measured using this approach is approximately three times as great as the variance that results when noise power is measured using a conventional noise figure meter.
- a noise measurement system measures noise power within one or more designated frequency bands of an applied signal.
- the noise measurement system includes a frequency converter that frequency translates the applied signal by a set of equally spaced frequencies to form a corresponding set of intermediate frequency signals.
- a sampler in the noise measurement system measures the noise in at least two measurement bands of each of the intermediate frequency signals that are separated by the frequency spacing of the equally spaced frequencies.
- the noise measurement system also includes a signal processor that determines the noise power in the designated frequency band of the applied signal based on the noise measurements by the sampler.
- Alternative embodiments of the present invention are directed to a noise measurement method.
- FIG. 1 shows a series of frequency bands within an exemplary noise spectrum.
- FIG. 2 shows a conventional noise figure meter.
- FIGS. 3A-3D show conventional direct conversion receivers and associated measurement noise spectra.
- FIGS. 4A-4D show alternative conventional noise measurement receivers and associated measurement noise spectra.
- FIG. 5 shows a noise measurement system according to embodiments of the present invention.
- FIGS. 6A-6D show noise spectra associated with the noise measurement system of FIG. 5 .
- FIG. 7 shows a noise measurement method according to alternative embodiments of the present invention.
- FIG. 8 shows a plot of measurement accuracy of the noise measurement system and noise measurement method according to the embodiments of the present invention.
- FIG. 6A shows an exemplary signal S IN with a series of frequency bands a-e designated within the spectrum of the signal S IN .
- a noise measurement system 10 and noise measurement method 20 according to embodiments of the present invention measure noise power within one or more of the frequency bands a-e.
- the noise measurement system 10 (shown in FIG. 5 ) includes a frequency converter 12 cascaded with a sampler 14 and a signal processor 16 .
- a display 18 or other output device optionally coupled to the signal processor 16 enables the noise power measured by the noise measurement system 10 to be displayed.
- the frequency converter 12 typically includes a preselector FIL, a mixer M, a local oscillator LO, and output filters HP, LP.
- the preselector FIL limits the bandwidth of the signal S IN prior to application to the mixer M to reduce the effect of higher order mixing products that cause unwanted noise sidebands to fall within the bandwidth of intermediate frequency signals IF 1 -IF n at the output of the frequency converter 12 .
- the mixer M is a double-balanced mixer.
- any other suitable frequency translation device is alternatively used in the frequency converter 12 .
- the local oscillator LO is any suitable signal source capable of providing a succession of local oscillator signals SLO 1 -SLO n to the mixer M, wherein each of the local oscillator signals has a frequency belonging to a designated set of equally-spaced frequencies f 1 -f n .
- the local oscillator LO is signal source that can be stepped in frequency, or a comb generator with an adjustable filter that provides the local oscillator signal SLO 1 -SLO n successively at each of the frequencies f 1 -f n .
- the output filter HP is a high pass filter that filters close-in noise from the local oscillator LO that gets translated by the mixer M into the frequency range of the intermediate frequency signals IF 1 -IF n .
- the output filter LP is typically a low pass filter that limits the bandwidth of each of the intermediate frequency signals IF 1 -IF n present at the output of the frequency converter 12 to prevent aliasing when the intermediate frequency signals are each sampled by the sampler 14 .
- the sampler 14 is an analog-to-digital (A-to-D) converter, signal digitizer or other type of sampling system suitable for sampling each of the intermediate frequency signals IF 1 -IF n at the output of the frequency converter 12 .
- the signal processor 16 in this example implemented with a digital signal processor (DSP), processes the samples of the intermediate frequency signals IF 1 -IF n acquired by the sampler 14 to determine the noise power in one or more of the frequency bands a-e within the signal S IN .
- DSP digital signal processor
- the frequency converter 12 frequency translates the signal S IN to provide the intermediate frequency signals IF 1 -IF N at the input to the sampler 14 .
- This frequency translation includes mixing the signal S IN with each of the local oscillator signals SLO 1 -SLO n .
- the local oscillator signals SLO 1 -SLO n having the corresponding frequencies f 1 -f n are separated by a frequency spacing f s .
- FIGS. 6B, 6C , 6 D show examples of three intermediate frequency signals IF 1 -IF n that result at the output of the frequency converter 12 when the signal S IN is translated by three equally-spaced frequencies f 1 -f 3 .
- the resulting intermediate frequency signals IF 1 -IF 3 each include a superposition of overlapping upper and lower spectral sidebands.
- the upper sidebands USB x and lower sidebands LSB x though overlapping, are shown vertically offset from each other in FIGS. 6B-6D .
- Noise of each of the intermediate frequency signals is measured in two or more measurement bands B 1 -B k defined within the spectrum of each of the intermediate frequency signals IF 1 -IF 3 .
- the measurement bands B 1 -B K are each separated by the frequency spacing f s that separates the frequencies f 1 -f n .
- the sampler 14 acquires samples of each of the intermediate frequency signals at a sufficiently high sampling rate to accommodate signal bandwidths that are wider than the frequency spacing f s of the local oscillator signals SLO 1 -SLO n .
- the sampling rate of the sampler 14 is suitably high to satisfy the Nyquist criteria for the two or more defined measurement bands BL-B K when separated by the frequency spacing f s .
- These measurement bands B 1 -B K are defined in the intermediate frequency signals IF 1 -IF n either by filtering the intermediate frequency signals prior to the sampling by the sampler 14 , or by digital filtering or other signal processing of the samples that are acquired of the intermediate frequency signals IF 1 -IF n by the sampler 14 .
- various frequency bands a-e of the signal S IN overlap within two or more of the measurement bands B 1 -B K .
- FIG. 6B shows the spectrum of the intermediate frequency signal IF 1 that results at a time t 1 when the local oscillator LO providing the local oscillator signal SLO 1 has a frequency f 1 .
- Noise components b 1 and b 2 of the frequency band b in the signal S IN are shown overlapping within a measurement band B 1 in the intermediate frequency signal IF 1 .
- Frequency bands a and c in the signal S IN overlap within the measurement band B 2 in the intermediate frequency signal IF 1 .
- Measuring the overlapping noise components b 1 , b 2 in the measurement band B 1 results in the noise measurement M b .
- the noise measurement M ac represents the sum of the noise power Na in the frequency band a translated to a lower sideband LSB 1 in the intermediate frequency signal IF 1 , and the noise power N c in the frequency band c translated to an overlapping upper sideband USB 1 of the intermediate frequency signal IF 1 .
- FIG. 6C shows the spectrum of the intermediate frequency signal IF 2 that results when the local oscillator LO providing the local oscillator signal SLO 2 has a frequency f 2 .
- Noise components c 1 and c 2 of the frequency band c in the signal S IN are shown overlapping within a measurement band B 1 in the intermediate frequency signal IF 2 .
- Frequency bands b and d in the signal S IN overlap within the measurement band B 2 in the intermediate frequency signal IF 2 .
- Frequency bands a and e in the signal S IN overlap within the measurement band B 3 in the intermediate frequency signal IF 2 .
- Measuring overlapping noise components c 1 , c 2 in the measurement band B 1 results in the noise measurement M c .
- the noise measurement M bd represents the sum of the noise power Nb in the frequency band b translated to a lower sideband LSB 2 in the intermediate frequency signal IF 2 , and the noise power N d in the frequency band d translated to an overlapping upper sideband USB 2 of the intermediate frequency signal IF 2 .
- the noise measurement M ae represents the sum of the noise power Na in the frequency band a translated to a lower sideband LSB 2 in the intermediate frequency signal IF 2 , and the noise power Ne in the frequency band e translated to an overlapping upper sideband USB 2 of the intermediate frequency signal IF 2 .
- FIG. 6D shows the spectrum of the intermediate frequency signal IF 3 that results when the local oscillator LO providing the local oscillator signal SLO 3 has a frequency f 3 .
- Noise components d 1 and d 2 of the frequency band d in the signal S IN are shown overlapping within a measurement band B 1 in the intermediate frequency signal IF 3 .
- Frequency bands c and e in the signal S IN overlap within the measurement band B 2 in the intermediate frequency signal IF 3 .
- Measuring the overlapping noise components d 1 , d 2 in the measurement band B 1 results in the noise measurement M d .
- the noise measurement M ce represents the sum of the noise power N c in the frequency band e translated to a lower sideband LSB 3 in the intermediate frequency signal IF 3 , and the noise power N e in the frequency band e translated to an overlapping upper sideband USB 3 of the intermediate frequency signal IF 3 .
- the noise power in one or more of the frequency bands a-e of the signal S IN is determined by the signal processor 16 based on the noise measurements.
- the noise power determination is facilitated by expressing the relationship between the noise measurements M b , M c , M d , M ac , M bd , M ae , M ce , and the noise powers N a , N b , N c , N d , N e in the frequency bands a-e, in a matrix equation 1.
- Equation 1 applies the least squares method, described for example in by Charles W. Therrien, Discrete Random Signals and Statistical Signal Processing , pages 518-523, ISBN 0-13-852112-3, herein incorporated by reference, to the matrix equation 1 leads to the solution for the noise power matrix [N] in equation 2, where [H] T is the transpose of the matrix [H].
- the noise power in one or more of the frequency bands a-e of the signal S IN is readily available.
- the noise power is determined within the frequency band c of the signal S IN for the purpose of illustration, the noise power can also be determined in other frequency bands within the signal S IN .
- the noise power (for example N c ) provides a measure of the statistics of the noise in the frequency band c.
- the measured noise expressed in the matrix [M] is also a random variable and the variance of the noise measurements in the matrix [M] provides a measure of uncertainty in the noise measurements.
- the variance of the noise measurement M b is 1.56
- the variance of the noise measurement M c is 1.35
- the variance of the noise measurement M d is 1.56, where each of these variances is normalized to the variance that results when a conventional noise figure meter is used to determine the noise power within a designated frequency band of the signal S IN .
- the noise power in this example has been determined based on the frequency translation of the signal S IN by three frequencies f 1 -f 3 for the purpose of illustration.
- the accuracy of the noise power measurements increases.
- FIG. 8 shows the resulting increase in measurement accuracy, depicted as a decrease in normalized measurement variance, as the number n of local oscillator frequencies f 1 -IF n increases.
- noise power is determined by a noise measurement method 20 shown in the flow diagram of FIG. 7 .
- the noise measurement method 20 which includes steps 22 - 26 , determines the noise power within one or more of the designated frequency bands a-e of the signal S IN .
- Step 22 includes frequency translating the signal S IN by the set of equally spaced frequencies f 1 -f n to form the intermediate frequency signals IF 1 -IF n frequency translating typically includes mixing the signal S IN with each of the local oscillator signals SLO 1 -SLO n .
- Step 24 includes measuring the noise in the two or more measurement bands B 1 -B K within each of the intermediate frequency signals IF 1 -IF n that are separated by the frequency spacing f s of the equally spaced frequencies f 1 -f n .
- the noise measurement includes acquiring samples of each of the intermediate frequency signals IF 1 -IF n at a sufficiently high sampling rate to accommodate signal bandwidths of the two or more defined measurement bands B 1 -B K that are separated by the frequency spacing f s .
- the measurement bands B 1 -B K are defined in the intermediate frequency signals IF 1 -IF n either by filtering the intermediate frequency signals prior to the sampling by the sampler 14 , or by digital filtering or other signal processing of the samples that are acquired of the intermediate frequency signals IF 1 -F n by the sampler 14 .
- noise power in the designated frequency band of the signal S IN is determined based on the noise measurements performed in step 24 .
- the noise power in one or more designated frequency bands of the signal S IN is determined by solution of the matrix equation 1 and by application of the least square method.
Abstract
Noise power is measured within one or more designated frequency bands of an applied signal. The measurement includes frequency translating the applied signal by a set of equally spaced frequencies to form a corresponding set of intermediate frequency signals, measuring the noise in at least two measurement bands of each of the intermediate frequency signals that are separated by the frequency spacing of the equally spaced frequencies, and determining the noise power in the designated frequency band of the applied signal based on the noise measurements.
Description
- One measure of the integrity of a communication system is the amount of noise present within the system. Characterizing the noise of a communication system involves measuring noise power within one or more specified frequency bands, shown for example in the exemplary noise spectrum of
FIG. 1 . Measuring noise power, in turn, involves measuring or estimating the statistics of the noise, which is inherently random. - A conventional noise figure meter (shown in
FIG. 2 ) provides accurate measurements of noise power within a designated frequency band. In the noise figure meter, noise within the frequency band is down-converted and measured by a noise power detector. Because the noise figure meter includes multiple frequency conversion stages, the noise figure meters are typically expensive to manufacture. - Conventional direct conversion receivers (shown in
FIGS. 3A-3B ) are also used to measure noise power. A direct conversion receiver is typically less expensive to manufacture than a noise figure meter because the receiver includes only a single frequency conversion stage. The direct conversion receiver converts the noise within a frequency band (for example, frequency band c shown inFIG. 3C ) to a baseband noise signal (shown inFIG. 3D ) that is measured by a noise power detector (as shown inFIG. 3A ) or by a narrow-band analog-to-digital converter (as shown inFIG. 3B ). However, these direct conversion receivers are not as accurate as noise figure meters. One measure of error, the variance of the measured noise power, is substantially higher for the direct conversion receiver than for the noise figure meter. For example, the variance of the noise power measured by a typical direct conversion receiver is approximately twice as great as the variance of the noise power measured by a noise figure meter. - An alternative approach to noise power measurement is shown in
FIG. 4A . In this approach, a frequency conversion stage causes upper and lower noise sidebands to overlap within a single measurement band B.FIGS. 4B-4D show these overlapping noise sidebands within the single measurement band B as a local oscillator within the frequency conversion stage is stepped in frequency between frequencies f1, f2 and f3. A noise power detector then measures the noise power in the single measurement band B, with the local oscillator at each of the stepped frequencies. While this approach can take advantage of low-cost signal processing to extract the noise power in a designated frequency band (for example, frequency band c) based on the overlapping noise sidebands, measurement accuracy is not as good as that of the conventional noise figure meter. For example, the variance of the noise power measured using this approach is approximately three times as great as the variance that results when noise power is measured using a conventional noise figure meter. - In view of the above, there is a need for an accurate noise measurement system that does not rely on the multiple frequency conversion stages of a noise figure meter.
- A noise measurement system according to embodiments of the present invention measures noise power within one or more designated frequency bands of an applied signal. The noise measurement system includes a frequency converter that frequency translates the applied signal by a set of equally spaced frequencies to form a corresponding set of intermediate frequency signals. A sampler in the noise measurement system measures the noise in at least two measurement bands of each of the intermediate frequency signals that are separated by the frequency spacing of the equally spaced frequencies. The noise measurement system also includes a signal processor that determines the noise power in the designated frequency band of the applied signal based on the noise measurements by the sampler. Alternative embodiments of the present invention are directed to a noise measurement method.
-
FIG. 1 shows a series of frequency bands within an exemplary noise spectrum. -
FIG. 2 shows a conventional noise figure meter. -
FIGS. 3A-3D show conventional direct conversion receivers and associated measurement noise spectra. -
FIGS. 4A-4D show alternative conventional noise measurement receivers and associated measurement noise spectra. -
FIG. 5 shows a noise measurement system according to embodiments of the present invention. -
FIGS. 6A-6D show noise spectra associated with the noise measurement system ofFIG. 5 . -
FIG. 7 shows a noise measurement method according to alternative embodiments of the present invention. -
FIG. 8 shows a plot of measurement accuracy of the noise measurement system and noise measurement method according to the embodiments of the present invention. -
FIG. 6A shows an exemplary signal SIN with a series of frequency bands a-e designated within the spectrum of the signal SIN. Anoise measurement system 10 andnoise measurement method 20 according to embodiments of the present invention measure noise power within one or more of the frequency bands a-e. The noise measurement system 10 (shown inFIG. 5 ) includes afrequency converter 12 cascaded with asampler 14 and asignal processor 16. Adisplay 18 or other output device optionally coupled to thesignal processor 16 enables the noise power measured by thenoise measurement system 10 to be displayed. - The
frequency converter 12 typically includes a preselector FIL, a mixer M, a local oscillator LO, and output filters HP, LP. The preselector FIL limits the bandwidth of the signal SIN prior to application to the mixer M to reduce the effect of higher order mixing products that cause unwanted noise sidebands to fall within the bandwidth of intermediate frequency signals IF1-IFn at the output of thefrequency converter 12. In this example, the mixer M is a double-balanced mixer. However, any other suitable frequency translation device is alternatively used in thefrequency converter 12. - The local oscillator LO is any suitable signal source capable of providing a succession of local oscillator signals SLO1-SLOn to the mixer M, wherein each of the local oscillator signals has a frequency belonging to a designated set of equally-spaced frequencies f1-fn. In a typical implementation, the local oscillator LO is signal source that can be stepped in frequency, or a comb generator with an adjustable filter that provides the local oscillator signal SLO1-SLOn successively at each of the frequencies f1-fn.
- The output filter HP is a high pass filter that filters close-in noise from the local oscillator LO that gets translated by the mixer M into the frequency range of the intermediate frequency signals IF1-IFn. The output filter LP is typically a low pass filter that limits the bandwidth of each of the intermediate frequency signals IF1-IFn present at the output of the
frequency converter 12 to prevent aliasing when the intermediate frequency signals are each sampled by thesampler 14. - The
sampler 14 is an analog-to-digital (A-to-D) converter, signal digitizer or other type of sampling system suitable for sampling each of the intermediate frequency signals IF1-IFn at the output of thefrequency converter 12. Thesignal processor 16, in this example implemented with a digital signal processor (DSP), processes the samples of the intermediate frequency signals IF1-IFn acquired by thesampler 14 to determine the noise power in one or more of the frequency bands a-e within the signal SIN. - To determine the noise power within a designated frequency band within the signal SIN, the
frequency converter 12 frequency translates the signal SIN to provide the intermediate frequency signals IF1-IFN at the input to thesampler 14. This frequency translation includes mixing the signal SIN with each of the local oscillator signals SLO1-SLOn. The local oscillator signals SLO1-SLOn, having the corresponding frequencies f1-fn are separated by a frequency spacing fs.FIGS. 6B, 6C , 6D show examples of three intermediate frequency signals IF1-IFn that result at the output of thefrequency converter 12 when the signal SIN is translated by three equally-spaced frequencies f1-f3. The resulting intermediate frequency signals IF1-IF3 each include a superposition of overlapping upper and lower spectral sidebands. The upper sidebands USBx and lower sidebands LSBx, though overlapping, are shown vertically offset from each other inFIGS. 6B-6D . - Noise of each of the intermediate frequency signals, for example intermediate frequency signals IF1-IF3, is measured in two or more measurement bands B1-Bk defined within the spectrum of each of the intermediate frequency signals IF1-IF3. The measurement bands B1-BK are each separated by the frequency spacing fs that separates the frequencies f1-fn. To measure the noise of the intermediate frequency signals, the
sampler 14 acquires samples of each of the intermediate frequency signals at a sufficiently high sampling rate to accommodate signal bandwidths that are wider than the frequency spacing fs of the local oscillator signals SLO1-SLOn. Particularly, the sampling rate of thesampler 14 is suitably high to satisfy the Nyquist criteria for the two or more defined measurement bands BL-BK when separated by the frequency spacing fs. These measurement bands B1-BK are defined in the intermediate frequency signals IF1-IFn either by filtering the intermediate frequency signals prior to the sampling by thesampler 14, or by digital filtering or other signal processing of the samples that are acquired of the intermediate frequency signals IF1-IFn by thesampler 14. Depending on the frequencies f1-fn of the local oscillator signals SLO1-SLOn, various frequency bands a-e of the signal SIN overlap within two or more of the measurement bands B1-BK. -
FIG. 6B shows the spectrum of the intermediate frequency signal IF1 that results at a time t1 when the local oscillator LO providing the local oscillator signal SLO1 has a frequency f1. Noise components b1 and b2 of the frequency band b in the signal SIN are shown overlapping within a measurement band B1 in the intermediate frequency signal IF1. Frequency bands a and c in the signal SIN overlap within the measurement band B2 in the intermediate frequency signal IF1. Measuring the overlapping noise components b1, b2 in the measurement band B1, results in the noise measurement Mb. Measuring noise of the overlapping frequency bands a and c in the measurement band B2 results in a noise measurement Mac=Na+Nc. The noise measurement Mac represents the sum of the noise power Na in the frequency band a translated to a lower sideband LSB1 in the intermediate frequency signal IF1, and the noise power Nc in the frequency band c translated to an overlapping upper sideband USB1 of the intermediate frequency signal IF1. -
FIG. 6C shows the spectrum of the intermediate frequency signal IF2 that results when the local oscillator LO providing the local oscillator signal SLO2 has a frequency f2. Noise components c1 and c2 of the frequency band c in the signal SIN are shown overlapping within a measurement band B1 in the intermediate frequency signal IF2. Frequency bands b and d in the signal SIN overlap within the measurement band B2 in the intermediate frequency signal IF2. Frequency bands a and e in the signal SIN overlap within the measurement band B3 in the intermediate frequency signal IF2. Measuring overlapping noise components c1, c2 in the measurement band B1 results in the noise measurement Mc. Measuring noise of the overlapping frequency bands b and d in the measurement band B2 results in a noise measurement Mbd=Nb+Nd. The noise measurement Mbd represents the sum of the noise power Nb in the frequency band b translated to a lower sideband LSB2 in the intermediate frequency signal IF2, and the noise power Nd in the frequency band d translated to an overlapping upper sideband USB2 of the intermediate frequency signal IF2. Measuring noise of the overlapping frequency bands a and e in the measurement band B3 results in a noise measurement Mae=Na+Ne. The noise measurement Mae represents the sum of the noise power Na in the frequency band a translated to a lower sideband LSB2 in the intermediate frequency signal IF2, and the noise power Ne in the frequency band e translated to an overlapping upper sideband USB2 of the intermediate frequency signal IF2. -
FIG. 6D shows the spectrum of the intermediate frequency signal IF3 that results when the local oscillator LO providing the local oscillator signal SLO3 has a frequency f3. Noise components d1 and d2 of the frequency band d in the signal SIN are shown overlapping within a measurement band B1 in the intermediate frequency signal IF3. Frequency bands c and e in the signal SIN overlap within the measurement band B2 in the intermediate frequency signal IF3. Measuring the overlapping noise components d1, d2 in the measurement band B1 results in the noise measurement Md. Measuring noise of the overlapping frequency bands c and e in the measurement band B2 results in a noise measurement Mce=Nc+Ne. The noise measurement Mce represents the sum of the noise power Nc in the frequency band e translated to a lower sideband LSB3 in the intermediate frequency signal IF3, and the noise power Ne in the frequency band e translated to an overlapping upper sideband USB3 of the intermediate frequency signal IF3. - The noise power in one or more of the frequency bands a-e of the signal SIN is determined by the
signal processor 16 based on the noise measurements. In this example, the noise power determination is facilitated by expressing the relationship between the noise measurements Mb, Mc, Md, Mac, Mbd, Mae, Mce, and the noise powers Na, Nb, Nc, Nd, Ne in the frequency bands a-e, in amatrix equation 1. - In an alternative notation,
Equation 1 is expressed by the matrix equation [H]*[N]=[M]. Applying the least squares method, described for example in by Charles W. Therrien, Discrete Random Signals and Statistical Signal Processing, pages 518-523, ISBN 0-13-852112-3, herein incorporated by reference, to thematrix equation 1 leads to the solution for the noise power matrix [N] inequation 2, where [H]T is the transpose of the matrix [H].
[N]={[H]T*[H]}-1*[H]T*[M] (2)
This solution for the noise power matrix [N] is expressed explicitly inequation 3 for the example where the set of frequencies f1-fn is made up of the three frequencies f1, f2, f3. - From
equation 3, the noise power in one or more of the frequency bands a-e of the signal SIN is readily available. For example, the noise power in the frequency band c of the signal SIN is expressed in terms of the measured powers as follows: - While the noise power is determined within the frequency band c of the signal SIN for the purpose of illustration, the noise power can also be determined in other frequency bands within the signal SIN.
- Due to the random nature of noise in the signal SIN, the noise power (for example Nc) provides a measure of the statistics of the noise in the frequency band c. The measured noise expressed in the matrix [M] is also a random variable and the variance of the noise measurements in the matrix [M] provides a measure of uncertainty in the noise measurements. In an exemplary noise measurement, where the noise measurements are performed with the local oscillator signals SLO1-SLO3 at the frequencies f1-f3, the variance of the noise measurement Mb is 1.56, the variance of the noise measurement Mc is 1.35 and the variance of the noise measurement Md is 1.56, where each of these variances is normalized to the variance that results when a conventional noise figure meter is used to determine the noise power within a designated frequency band of the signal SIN.
- The noise power in this example has been determined based on the frequency translation of the signal SIN by three frequencies f1-f3 for the purpose of illustration. When more than the three local oscillator frequencies f1-f3 are used in the frequency translations and when frequency bands in addition to the frequency bands a-e are designated in the signal SIN, the accuracy of the noise power measurements increases.
FIG. 8 shows the resulting increase in measurement accuracy, depicted as a decrease in normalized measurement variance, as the number n of local oscillator frequencies f1-IFn increases. - According to alternative embodiments of the present invention, noise power is determined by a
noise measurement method 20 shown in the flow diagram ofFIG. 7 . Thenoise measurement method 20, which includes steps 22-26, determines the noise power within one or more of the designated frequency bands a-e of the signal SIN. Step 22 includes frequency translating the signal SIN by the set of equally spaced frequencies f1-fn to form the intermediate frequency signals IF1-IFn frequency translating typically includes mixing the signal SIN with each of the local oscillator signals SLO1-SLOn. -
Step 24 includes measuring the noise in the two or more measurement bands B1-BK within each of the intermediate frequency signals IF1-IFn that are separated by the frequency spacing fs of the equally spaced frequencies f1-fn. The noise measurement includes acquiring samples of each of the intermediate frequency signals IF1-IFn at a sufficiently high sampling rate to accommodate signal bandwidths of the two or more defined measurement bands B1-BK that are separated by the frequency spacing fs. The measurement bands B1-BK are defined in the intermediate frequency signals IF1-IFn either by filtering the intermediate frequency signals prior to the sampling by thesampler 14, or by digital filtering or other signal processing of the samples that are acquired of the intermediate frequency signals IF1-Fn by thesampler 14. Instep 26, noise power in the designated frequency band of the signal SIN is determined based on the noise measurements performed instep 24. In one example, the noise power in one or more designated frequency bands of the signal SIN is determined by solution of thematrix equation 1 and by application of the least square method. - While the embodiments of the present invention have been illustrated in detail, it should be apparent that modifications and adaptations to these embodiments may occur to one skilled in the art without departing from the scope of the present invention as set forth in the following claims.
Claims (20)
1. A system for measuring noise within a designated frequency band of an applied signal, comprising:
a frequency converter frequency translating the applied signal by a set of equally spaced frequencies to provide a corresponding set of intermediate frequency signals;
a sampler measuring the noise in at least two measurement bands of each of the intermediate frequency signals that are separated by the frequency spacing of the equally spaced frequencies; and
a signal processor determining the noise power in the frequency band of the applied signal based on the noise measured by the sampler.
2. The system of claim 1 wherein the frequency converter includes a preselector limiting the bandwidth of the applied signal, a mixer having a first port coupled to the preselector, a cascaded high pass filter and a low pass filter interposed between a second port of the mixer and the sampler, wherein a third port of the mixer receives a local oscillator signal having at successive times the frequencies in the set of equally spaced frequencies.
3. The system of claim 1 wherein measuring the noise in the at least two measurement bands of each of the intermediate frequency signals that are separated by the frequency spacing of the equally spaced frequencies includes sampling each of the intermediate frequency signals.
4. The system of claim 3 wherein the at least two measurement bands are defined by filtering each of the intermediate frequency signals prior to sampling each of the intermediate frequency signals.
5. The system of claim 3 wherein the at least two measurement bands of each of the intermediate frequency signals are defined by sampling each of the intermediate frequency signals and digitally filtering samples acquired from the sampling.
6. The system of claim 1 wherein the signal processor determines the noise power in the frequency band of the applied signal by solving a matrix equation resulting from the noise measured in the at least two measurement bands of each of the intermediate frequency signals by the sampler, and applying the least squares method.
7. The system of claim 1 wherein the at least two measurement bands of each of the intermediate frequency signals include an upper sideband and a lower sideband that are overlapping.
8. The system of claim 1 wherein frequency translating the applied signal by the set of equally spaced frequencies to provide the corresponding set of intermediate frequency signals includes mixing the applied signal with a local oscillator signal having at successive times the frequencies in the set of equally spaced frequencies.
9. The system of claim 3 wherein the at least two measurement bands of each of the intermediate frequency signals include an upper sideband and a lower sideband that are overlapping.
10. A method for measuring noise within a frequency band of an applied signal, comprising:
(a) frequency translating the applied signal by a set of equally spaced frequencies to form a corresponding set of intermediate frequency signals;
(b) measuring the noise in at least two measurement bands of each of the intermediate frequency signals that are separated by the frequency spacing of the equally spaced frequencies; and
(c) determining the noise power in the frequency band of the applied signal based on the noise measured in (b).
11. The method of claim 10 wherein (a) includes mixing the applied signal with a local oscillator signal having at successive times the frequencies in the set of equally spaced frequencies.
12. The method of claim 10 wherein (b) includes sampling each of the intermediate frequency signals.
13. The method of claim 11 wherein (b) includes sampling each of the intermediate frequency signals.
14. The method of claim 12 wherein the at least two measurement bands are defined by filtering each of the intermediate frequency signals prior to sampling each of the intermediate frequency signals.
15. The method of claim 13 wherein the at least two measurement bands are defined by filtering each of the intermediate frequency signals prior to sampling each of the intermediate frequency signals.
16. The method of claim 10 wherein the at least two measurement bands are defined by sampling each of the intermediate frequency signals and digitally filtering samples acquired from the sampling.
17. The method of claim 10 wherein the at least two measurement bands of each of the intermediate frequency signals are defined by filtering the intermediate frequency signals in the set.
18. The method of claim 10 wherein (c) includes solving a matrix equation resulting from the noise measurements in (b) and applying the least squares method.
19. The method of claim 11 wherein (c) includes solving a matrix equation resulting from the noise measurements in (b) and applying the least squares method.
20. The method of claim 10 wherein the at least two measurement bands in each of the intermediate frequency signals each include an upper sideband and a lower sideband that are overlapping.
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/699,526 US7010443B2 (en) | 2003-10-31 | 2003-10-31 | Noise measurement system and method |
DE102004034764A DE102004034764A1 (en) | 2003-10-31 | 2004-07-19 | Noise measurement system and method |
JP2004313708A JP2005134398A (en) | 2003-10-31 | 2004-10-28 | Noise measurement system and method |
GB0423956A GB2407875B (en) | 2003-10-31 | 2004-10-28 | Noise measurement system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/699,526 US7010443B2 (en) | 2003-10-31 | 2003-10-31 | Noise measurement system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
US20050096859A1 true US20050096859A1 (en) | 2005-05-05 |
US7010443B2 US7010443B2 (en) | 2006-03-07 |
Family
ID=33518220
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/699,526 Expired - Fee Related US7010443B2 (en) | 2003-10-31 | 2003-10-31 | Noise measurement system and method |
Country Status (4)
Country | Link |
---|---|
US (1) | US7010443B2 (en) |
JP (1) | JP2005134398A (en) |
DE (1) | DE102004034764A1 (en) |
GB (1) | GB2407875B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110178776A1 (en) * | 2008-05-28 | 2011-07-21 | Advantest Corporation | Excessive noise ratio deriving device, noise figure deriving device, method, program, and recording medium |
CN104360178A (en) * | 2014-10-30 | 2015-02-18 | 东北电力大学 | Method for measuring low-frequency noise of inverters and diagnosing faults thereof |
US20150369898A1 (en) * | 2014-06-24 | 2015-12-24 | Tektronix, Inc. | Phase noise correction system for discrete time signal processing |
US20190190622A1 (en) * | 2017-12-20 | 2019-06-20 | Samsung Electronics Co., Ltd. | Wireless communication apparatuses performing selective noise filtering and methods of operating the same |
US11029388B2 (en) * | 2018-01-29 | 2021-06-08 | Infineon Technologies Ag | Spectral estimation of noise in radar apparatus |
CN115963333A (en) * | 2023-03-16 | 2023-04-14 | 国仪量子(合肥)技术有限公司 | Noise spectral density scanning method, scanning device, storage medium, and electronic apparatus |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3962785B2 (en) * | 2003-07-02 | 2007-08-22 | テクトロニクス・インターナショナル・セールス・ゲーエムベーハー | Signal analyzer and frequency domain data generation method |
US7446692B2 (en) * | 2007-03-30 | 2008-11-04 | Microtune (Texas), L.P. | Digital radio system and method of operation |
KR101061513B1 (en) | 2009-12-21 | 2011-09-02 | 한국전자통신연구원 | Noise Figure Measurement Apparatus and Method |
US8972215B2 (en) * | 2011-03-30 | 2015-03-03 | Mitsubishi Electric Research Laboratories, Inc. | Method and system for determining parameters of sinusoidal signals |
JP6547648B2 (en) * | 2016-02-04 | 2019-07-24 | 富士通株式会社 | Optical signal-to-noise ratio measuring apparatus and optical signal-to-noise ratio measuring method |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4864218A (en) * | 1988-11-09 | 1989-09-05 | Cascade Microtech, Inc. | Method of compensating for frequency errors in noise power meters |
US5416422A (en) * | 1994-05-20 | 1995-05-16 | Hewlett-Packard Company | Apparatus and method for determining single sideband noise figure from double sideband measurements |
US5619536A (en) * | 1993-03-15 | 1997-04-08 | Alcatel Radiotelephone | Digital superheterodyne receiver and baseband filter method used therein |
US5638034A (en) * | 1992-12-30 | 1997-06-10 | Nokia Telecommunications Oy | Method and arrangement for tuning a combiner filter |
US5926513A (en) * | 1997-01-27 | 1999-07-20 | Alcatel Alsthom Compagnie Generale D'electricite | Receiver with analog and digital channel selectivity |
US20030034857A1 (en) * | 2001-08-14 | 2003-02-20 | Prime Electronics & Statellitcs Inc. | Multiplexer using ceramic filter as stacked satellite down converter |
US20040039540A1 (en) * | 2000-10-17 | 2004-02-26 | Takeshi Nagasaka | Noise measuring device and method, recording medium |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE1938795A1 (en) * | 1968-08-09 | 1970-02-12 | Int Standard Electric Corp | Circuit arrangement for measuring the signal to noise ratio |
GB2234874B (en) * | 1981-08-17 | 1991-06-26 | Philips Electronic Associated | Electronic surveillance systems. |
US6229316B1 (en) * | 1995-09-08 | 2001-05-08 | Advantest Corporation | Measuring method by spectrum analyzer |
-
2003
- 2003-10-31 US US10/699,526 patent/US7010443B2/en not_active Expired - Fee Related
-
2004
- 2004-07-19 DE DE102004034764A patent/DE102004034764A1/en not_active Withdrawn
- 2004-10-28 JP JP2004313708A patent/JP2005134398A/en active Pending
- 2004-10-28 GB GB0423956A patent/GB2407875B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4864218A (en) * | 1988-11-09 | 1989-09-05 | Cascade Microtech, Inc. | Method of compensating for frequency errors in noise power meters |
US5638034A (en) * | 1992-12-30 | 1997-06-10 | Nokia Telecommunications Oy | Method and arrangement for tuning a combiner filter |
US5619536A (en) * | 1993-03-15 | 1997-04-08 | Alcatel Radiotelephone | Digital superheterodyne receiver and baseband filter method used therein |
US5416422A (en) * | 1994-05-20 | 1995-05-16 | Hewlett-Packard Company | Apparatus and method for determining single sideband noise figure from double sideband measurements |
US5926513A (en) * | 1997-01-27 | 1999-07-20 | Alcatel Alsthom Compagnie Generale D'electricite | Receiver with analog and digital channel selectivity |
US20040039540A1 (en) * | 2000-10-17 | 2004-02-26 | Takeshi Nagasaka | Noise measuring device and method, recording medium |
US20030034857A1 (en) * | 2001-08-14 | 2003-02-20 | Prime Electronics & Statellitcs Inc. | Multiplexer using ceramic filter as stacked satellite down converter |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110178776A1 (en) * | 2008-05-28 | 2011-07-21 | Advantest Corporation | Excessive noise ratio deriving device, noise figure deriving device, method, program, and recording medium |
US8686741B2 (en) | 2008-05-28 | 2014-04-01 | Advantest Corporation | Excessive noise ratio deriving device, noise figure deriving device, method, program, and recording medium |
US20150369898A1 (en) * | 2014-06-24 | 2015-12-24 | Tektronix, Inc. | Phase noise correction system for discrete time signal processing |
CN105203860A (en) * | 2014-06-24 | 2015-12-30 | 特克特朗尼克公司 | Phase noise correction system for discrete time signal processing |
US9651646B2 (en) * | 2014-06-24 | 2017-05-16 | Tektronix, Inc. | Phase noise correction system for discrete time signal processing |
CN104360178A (en) * | 2014-10-30 | 2015-02-18 | 东北电力大学 | Method for measuring low-frequency noise of inverters and diagnosing faults thereof |
US20190190622A1 (en) * | 2017-12-20 | 2019-06-20 | Samsung Electronics Co., Ltd. | Wireless communication apparatuses performing selective noise filtering and methods of operating the same |
CN109951198A (en) * | 2017-12-20 | 2019-06-28 | 三星电子株式会社 | The method for executing the wireless telecom equipment of selective noise filtering and operating the equipment |
US10630394B2 (en) * | 2017-12-20 | 2020-04-21 | Samsung Electronics Co., Ltd. | Wireless communication apparatuses performing selective noise filtering and methods of operating the same |
US11029388B2 (en) * | 2018-01-29 | 2021-06-08 | Infineon Technologies Ag | Spectral estimation of noise in radar apparatus |
CN115963333A (en) * | 2023-03-16 | 2023-04-14 | 国仪量子(合肥)技术有限公司 | Noise spectral density scanning method, scanning device, storage medium, and electronic apparatus |
Also Published As
Publication number | Publication date |
---|---|
GB0423956D0 (en) | 2004-12-01 |
JP2005134398A (en) | 2005-05-26 |
DE102004034764A1 (en) | 2005-06-09 |
GB2407875A (en) | 2005-05-11 |
US7010443B2 (en) | 2006-03-07 |
GB2407875B (en) | 2007-02-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7688058B2 (en) | Integrated spectrum analyzer circuits and methods for providing on-chip diagnostics | |
US7010443B2 (en) | Noise measurement system and method | |
JPH08248070A (en) | Frequency spectrum analyzer | |
US8805313B2 (en) | Magnitude and phase response calibration of receivers | |
CN106405235A (en) | Spectrum analyzer and data processing method thereof | |
Al Adnani et al. | Spectrum analyzers today and tomorrow: part 1 towards filterbanks-enabled real-time spectrum analysis | |
US20230037910A1 (en) | Spectrum analyzer and method of controlling the same | |
US9450598B2 (en) | Two-stage digital down-conversion of RF pulses | |
CN106353594B (en) | A kind of fast multiresolution frequency spectrum analysis system and method | |
US20020168035A1 (en) | Parallel correlator archtitecture | |
JP4408967B2 (en) | Spectrum analyzer | |
US20080222228A1 (en) | Bank of cascadable digital filters, and reception circuit including such a bank of cascaded filters | |
US7272521B1 (en) | Measurement of phase nonlinearity of non-linear devices | |
US9877326B2 (en) | Measuring device and a measuring method for determining an active channel | |
US20220390495A1 (en) | Novel spectral analysis method based on digital pulse compression and chirp transform | |
JP5947943B1 (en) | Signal analysis apparatus and method | |
US9350470B1 (en) | Phase slope reference adapted for use in wideband phase spectrum measurements | |
US20050207512A1 (en) | Multichannel simultaneous real time spectrum analysis with offset frequency trigger | |
JP2019158418A (en) | Measuring system and measuring method | |
Braun | Spectrum analysis and EMI measurements based on time-domain methods | |
Schlecht et al. | Effects of local oscillator phase noise on submillimeter-wave spectrometer performance | |
RU16682U1 (en) | SELECTIVE SIGNAL MEASUREMENT DEVICE | |
Shenghua et al. | Evaluation and measurement of digital receiver for DBF radar | |
Grenkov et al. | A digital signal converter for radio astronomical systems | |
EP2141817A2 (en) | Noise measurement for radio squelch function |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: AGILENT TECHNOLOGIES, INC., COLORADO Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:TORIN, SHIGETSUNE;WONG, KENNETH H.;BLACKHAM, DAVID VERNON;REEL/FRAME:014322/0593;SIGNING DATES FROM 20031028 TO 20031029 |
|
CC | Certificate of correction | ||
REMI | Maintenance fee reminder mailed | ||
LAPS | Lapse for failure to pay maintenance fees | ||
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Expired due to failure to pay maintenance fee |
Effective date: 20100307 |